Marusan Securities Co., Ltd. (8613.T) - Deep Dive Research Report

Prepared 2026-06-13. All figures from company disclosures (決算短信 / Tanshin), the company website, irbank, and Japanese financial media. Marusan reports on a 1 April - 31 March fiscal year; "FY2026" means the year ended 31 March 2026.

A note on reporting cadence and the "concall" requirement

Marusan is a roughly ¥-billion-revenue, ~1,100-employee retail broker listed on the TSE Prime market. Like most Japanese small-caps it does not hold English-language earnings calls and does not publish call transcripts. Its public reporting consists of:

• Quarterly Tanshin (決算短信) financial summaries - Q1 (filed late July), H1/Q2 (late October), Q3 (late January, with a flash 速報値 in mid-January), and full-year (late April).
• A twice-yearly management-strategy briefing (決算・経営戦略説明会) with a slide deck, in May (full-year) and November (interim).

The five most recent reporting periods, which I treat as the concall-equivalent throughout, are:

Period 1 (28 Apr 2026) is ~46 days before today, so it satisfies the "within 90 days" requirement.

Section 1: What the company does

Marusan Securities is an independent, mid-sized Japanese brokerage that sells stocks, bonds and (above all) investment trusts (mutual funds) to individual Japanese savers, almost entirely face-to-face. It is not a trading house betting its own balance sheet, and it is not an online discount broker. It is a relationship business: a sales person sits across a table (or increasingly, on a screen) from a retail client - often an older, asset-rich Japanese household - and advises them on what to buy, earning a commission when they transact and a recurring trail fee for as long as they hold a fund.

That is the entire business. In its filings Marusan describes itself as a single segment: "investment and financial services centering on securities." Commissions received (受入手数料) make up roughly 99% of revenue. There is almost no proprietary trading P&L and almost no investment-banking franchise of consequence.

The founding story still explains the firm. Marusan traces to January 1910, when a securities shop called Marusan Tada Iwakichi Shoten opened. Over the next three decades the house name changed with its proprietors - Tada Iwakichi Shoten, then Yoshida Seishiro Shoten, then Nagao Hidekazu Shoten. The Nagao lineage matters today: the Nagao Natural Environment Foundation is the second-largest shareholder (~7.2%), a direct thread from the founding family to the modern register. In 1944 the firm absorbed Irisa Securities and took the name Marusan Securities; in 1949 it launched its in-house research letter, the Marusan Report; it listed on the Tokyo and Osaka second sections in 1986 and now sits on TSE Prime. The continuity is the point - this is a 115-year-old advisory house that has always made its living from giving retail investors a human to talk to.

The core value proposition. Marusan sells trust and advice to Japanese households who do not want to pick funds and stocks alone. Japan has spent two decades trying to move roughly half its ¥2-quadrillion of household financial assets out of zero-yield cash deposits and into securities (the "貯蓄から投資へ" - "from savings to investment" - policy push, now turbo-charged by the expanded NISA tax-free wrapper). The online brokers (SBI, Rakuten) win the self-directed, fee-sensitive, younger investor. Marusan wins the investor who wants someone to call - typically older, with more assets and less inclination to manage them. The product it is really selling is reassurance and curation, priced as commission.

How it actually works, concretely. A retail client walks into one of Marusan's ~29 branch locations (or is visited by a salesperson). The salesperson, armed with the house Marusan Report and a recommended-fund and recommended-stock list, proposes a portfolio - say a global-equity balanced fund and a basket of large-cap Japanese names (semiconductors, megabanks, heavy industry are recurring house themes). When the client buys the fund, Marusan books a one-time solicitation/distribution commission (募集手数料). For as long as the client holds that fund, Marusan books a recurring trust fee / trail commission (信託報酬) from the asset manager - a slice of the fund's annual management fee, paid in proportion to the assets Marusan has under custody. When the client trades a stock, Marusan books a brokerage commission (委託手数料). The strategic prize, repeated in every recent disclosure, is to grow the recurring trail-fee book big enough that it covers the firm's fixed costs regardless of how active markets are - converting a cyclical commission business into something closer to an annuity.

Management's framing: the firm targets growing investment-trust trail fees until they cover 55% of selling, general and administrative expenses by FY2028 - a stability metric, not a growth metric (medium-term plan, 26 Apr 2024).

Section 2: Business segments

Marusan reports as a single segment ("securities and financial-instruments business"), so there is no formal multi-segment P&L to dissect. But the commission line is the firm's true operating anatomy, and management runs the business along three commission pillars plus a financial-income tail. I treat those as de facto segments below, because they have genuinely different economics, customer behaviour and competitive exposure.

2.1 Investment trusts (the recurring annuity) - the strategic core

This is the largest and most strategically important pillar. It has two revenue streams: a one-time solicitation/distribution commission when a client buys a fund, and a recurring trust fee (trail commission) paid by the asset manager for as long as the client holds it. In FY2025 the trust-fee line (~¥7.6bn) was Marusan's single biggest commission category and grew ~15% year on year; solicitation commissions (~¥5.3bn) grew ~12%.

• Core capability: a distribution shelf of curated funds (balanced funds, global-equity strategies, high-dividend developed-market funds) and a salesforce trained to place them into retail portfolios and keep them there. The trail-fee book is the asset that took decades to build - it is a function of cumulative assets under custody, not this quarter's selling.
• Why it matters as a unit: it is the only part of Marusan whose revenue does not require a transaction. Management's entire medium-term thesis is to enlarge this book until it de-risks the whole firm. The plan targets a cumulative ¥300bn net inflow into equity investment trusts by FY2028.
• Competitive position: competes against every other distributor for shelf space and client wallet - the megabanks, Nomura/Daiwa, the post office, and increasingly the online platforms that sell no-load funds at zero front-end fee. Marusan's edge is the advised relationship; its exposure is that the front-load solicitation fee is exactly the line online brokers have driven to zero.
• Fit in the group: the growth bet and the stability engine simultaneously.

2.2 Equities / brokerage (the cyclical engine)

Stock brokerage commissions (委託手数料, ~¥5.4bn in FY2025) are earned when clients buy and sell Japanese equities on Marusan's recommendation. The house pushes thematic large-caps - semiconductors, megabanks, heavy-industry names - through its recommended-stock list.

• Core capability: the Marusan Report research voice and a salesforce that converts house calls into client orders; a small equity-underwriting/IPO function (Marusan acted in the syndicate for ~19 new and 2 secondary listings in FY2025).
• Why separate: purely transactional and the most market-sensitive line in the firm - it swung -21% in FY2025 (a soft tape year) and then +36% in FY2026 (a buoyant one). This is the line that makes Marusan a leveraged play on Japanese retail trading appetite.
• Competitive position: structurally the most exposed to online-broker fee compression; Marusan defends it with advice, not price.

2.3 Bonds and underwriting (the small, steady tail)

Bond underwriting and distribution is a minor line (~¥0.13bn in FY2025, +28% as corporate-bond issuance picked up; distribution volume ~¥32.7bn). It rounds out the product shelf - clients who want yield without equity risk - but is not a profit driver. Marusan is a distributor of others' paper, not a debt-capital-markets house.

2.4 Financial income and "everything else"

Net interest/financial income (margin lending, deposits) and ancillary services (safe-deposit boxes, gold bullion, custody, insurance agency via the group, securities lending) form the residual. The firm runs a very high equity ratio (~66%) and a net-cash balance sheet, so financial income is real but small. In April 2025 Marusan absorbed its subsidiary Marusan Finance by merger, consolidating the financing function into the parent.

Section 3: Products and business detail

The product catalogue is a retail shelf, not a factory. Marusan's "products" are the financial instruments it places into household portfolios:

• Investment trusts (mutual funds): the flagship shelf. House-promoted strategies include balanced funds, global-equity funds, developed-market high-dividend equity funds (e.g. a quarterly-distribution developed-market high-dividend fund) and small/mid-cap "blue-chip" open funds. Funds are sold both through branches and through the MARUSAN-NET online channel and a fund-wrap (discretionary) service. These generate the front-load solicitation fee and the recurring trail.
• Japanese equities: cash equities and margin (信用) trading, executed on house recommendations skewed to large-cap themes (semiconductors, megabanks, heavy industry). Includes a modest IPO allocation business (Marusan appears in underwriting syndicates).
• Bonds: Japanese government and corporate bonds, distributed to yield-seeking retail clients; a small corporate-bond underwriting role.
• Ancillary services: safe-deposit boxes, gold bullion trading, custody/safekeeping, securities lending, life- and non-life-insurance agency (through group entities), and stock-transfer agency services.

The "manufacturing" process is distribution and advice. Marusan does not manufacture funds (asset managers do). Its production process is: (1) the research function publishes the Marusan Report and maintains recommended-fund and recommended-stock lists; (2) the branch salesforce converts those views into client portfolios face-to-face; (3) the back office settles, custodies and reports. The binding constraints are the size and productivity of the salesforce, the branch footprint, and the cumulative assets under custody that drive trail fees. Digital "investment navigation" tools (online fund selection, MARUSAN-NET) are being layered on to support - not replace - the advisory model.

Geography. Marusan is domestic Japan only. As of end-March 2026 it operates a face-to-face network of 29 locations anchored on its Tokyo (Kojimachi, Chiyoda-ku) head office, with branches across Japan. There is no meaningful overseas business. Branch relocations (e.g. a September 2025 relocation) are part of an ongoing footprint optimisation rather than expansion.

Milestones that shaped the business: founding (1910); the name change to Marusan and the Irisa acquisition (1944); the launch of the Marusan Report research letter (1949); TSE/OSE second-section listing (1986); the FY2024 medium-term plan that reoriented the firm around recurring trail fees and capital return (Apr 2024); and the absorption of Marusan Finance (Apr 2025).

Section 4: Customers

Who buys. Marusan's customers are Japanese individuals - retail savers, skewed toward older, asset-rich households who prefer an adviser to a screen. The firm reports tens of thousands of shareholders and a retail-dominated client base; it is explicitly not an institutional or wholesale broker. There is no single dominant corporate account; revenue is spread across a long tail of individual clients, which is itself a form of diversification.

Who decides, and on what criteria. The buying "decision-maker" is the household itself, and the decision is driven less by price than by trust in the salesperson and the house view. Criteria are: do I believe this adviser, does the Marusan Report sound credible, and is my money safe with a 115-year-old firm carrying a ~66% equity ratio and net cash. The sales cycle is relationship-paced - it can take years to win a meaningful share of a wealthy client's wallet, and the relationship, once won, is sticky.

Why they choose Marusan. Advice and hand-holding for people who will not self-direct. An online broker is cheaper but gives you nothing to lean on; a megabank is large but impersonal. Marusan's pitch is a named human, a house research voice, and longevity.

Switching costs. Modest but real, and behavioural rather than contractual. There is no qualification testing or regulatory lock-in as in an industrial supplier. But the target client - older, advice-reliant, holding funds that already pay Marusan a trail - is inert by disposition. The same inertia that makes the trail-fee book valuable also makes these clients unlikely to move to a cheaper online platform.

Concentration. Low single-account concentration (a retail book), but high concentration in a single demographic (ageing Japanese households) and a single market (Japan). The risk is not "we lost our biggest customer" - it is "our customer base is ageing and the next generation self-directs online."

Contract structure / revenue predictability. Two distinct profiles. The trail-fee book is the closest thing to a recurring, contract-like revenue stream - it persists as long as clients hold funds, and it is the line management is deliberately growing to make the firm predictable. Brokerage and solicitation are transactional and discretionary, swinging with market mood (the FY2025 -21% / FY2026 +36% brokerage swing shows the amplitude). The whole strategic project is to shift the mix toward the predictable line.

Section 5: Competitive landscape

Marusan sits in an uncomfortable middle of the Japanese retail-brokerage industry, squeezed from above by scale players and from below by zero-fee online platforms.

The structure of the industry. Japanese retail securities distribution has three tiers: (1) the mega-brokers (Nomura, Daiwa) and the bank-affiliated houses (SMBC Nikko, Mizuho, Mitsubishi UFJ Morgan Stanley) with national scale and full investment banks attached; (2) the online discount brokers (SBI, Rakuten, Monex, Matsui) that have driven Japanese equity brokerage commissions toward zero and dominate the self-directed and younger segments; and (3) a shrinking tier of independent/regional face-to-face brokers - Marusan, Tokai Tokyo, Iwai Cosmo, Toyo, Aizawa, Mito and similar - that live on advised retail relationships.

Where Marusan wins and loses. It wins where advice and trust beat price: older, asset-rich clients who will not self-direct and value a 115-year-old name with a fortress balance sheet. It loses, structurally, on price: the online brokers have made the front-load solicitation fee and the equity-brokerage commission - two of Marusan's three pillars - free or near-free for self-directed investors. Marusan cannot win a price war and does not try to; its defence is the relationship and the recurring trail book.

Barriers to entry. Real but not high. A securities licence, a salesforce and a distribution shelf are not trivial to assemble, and a sticky trail-fee book takes years to grow. But there is no patent, no process secret, no regulatory moat that stops a better-capitalised or lower-cost rival from competing for the same clients - and the online brokers already are. The genuine barrier protecting Marusan is the inertia of its existing client base, not any structural barrier protecting new client acquisition.

Structural shifts. Three are reshaping the field: the expanded NISA tax-free regime (2024 onward) is enlarging the retail-investing pie but disproportionately routing new money to low-cost online platforms; fee compression continues to erode transactional commissions; and consolidation among regional/independent brokers is a recurring theme (e.g. the Tokai Tokyo group's expansion). Marusan's relevance to consolidation cuts both ways - its strong balance sheet and clean register make it a plausible consolidator or a plausible target.

Market caps are approximate peer-size references with currency and as-of date; they move daily and are not used to value Marusan.

The honest read: this is a competitive, commoditising distribution business with no durable moat beyond an ageing, sticky client base and a strong balance sheet. The moat is the inertia of the existing book, and it erodes with the demographics of that book.

Section 6: Industry

Demand drivers. Marusan's fortunes ride three forces. First, Japanese household asset reallocation - the multi-decade "savings to investment" push to move roughly half of ~¥2-quadrillion household financial assets out of cash and into securities, now accelerated by the 2024 NISA expansion. Second, the level and direction of the Japanese stock market - brokerage and solicitation commissions rise and fall with retail trading appetite, which tracks the Nikkei. Third, assets under custody - the recurring trail-fee line grows with cumulative AUM, which is a function of net inflows plus market appreciation.

Size and trajectory. The retail-securities distribution market in Japan is large and structurally growing in assets (NISA inflows, an equity bull market through 2024-2026, the BOJ's exit from negative rates restoring some yield to the system), even as unit economics on transactions compress. The pie is getting bigger; the margin per transaction is getting thinner; the winners are those who capture recurring asset-based revenue rather than per-trade fees - which is precisely why Marusan's strategy targets trail fees.

Where Marusan sits in the chain. It is the distribution end of the asset-management value chain: asset managers manufacture funds; exchanges and clearers handle execution and settlement; Marusan is the last mile to the retail saver, taking a slice for advice and access. It captures none of the manufacturing economics and all of the relationship economics.

Regulation. A heavily regulated industry: Marusan operates under the Financial Instruments and Exchange Act, supervised by the Kanto Local Finance Bureau and the FSA, with fiduciary-conduct ("customer-first") rules increasingly scrutinising whether advised sales serve the client (a real constraint on front-load-fee selling). The TSE's 2023-onward "cost-of-capital and share-price-conscious management" campaign directly shaped Marusan's capital-return policy.

Cyclicality. High, and concentrated in the transactional lines. Brokerage and solicitation revenue is geared to market mood; the FY2025 (-21% brokerage) versus FY2026 (+36% brokerage) swing is the cycle in two data points. The trail-fee line is the counter-cyclical ballast - it falls only when AUM falls, which is gentler than the swing in trading volumes. The entire industry behaves like a leveraged play on retail risk appetite, and Marusan is a fairly pure expression of it.

Tailwinds: NISA-driven inflows, demographic wealth concentration in advice-seeking older households, rising domestic rates restoring financial income. Headwinds: zero-fee online competition, fiduciary-conduct pressure on front-load fees, and the long-run demographic risk that the next generation self-directs.

Section 7: Growth triggers

Source basis: Marusan publishes no earnings-call transcript. The triggers below are drawn from the medium-term plan disclosure, the Tanshin filings, and the management-strategy briefing decks across the five most recent reporting periods, each dated.

• Cumulative ¥300bn net inflow into equity investment trusts by FY2028 (medium-term plan, 26 Apr 2024; reiterated through FY2026 reporting). This is the central growth target - growing the recurring trail-fee book.

  Progress flagged at the interim stage: equity investment-trust net inflows were running ahead of plan and the trust-fee-to-SG&A coverage ratio had reached 49.0% against the 55% target (H1 FY2025 briefing, ~Nov 2024).

• Cumulative ¥100bn net inflow into recommended Japanese stocks by FY2028 (medium-term plan, 26 Apr 2024). Management has repeatedly flagged this line running well ahead of its annualised pace (the Japanese-stock target was reported at ~168% of its 12-month goal in interim disclosure).

• Trust fees to cover 55% of SG&A by FY2028 (medium-term plan, 26 Apr 2024; repeated FY2025 and FY2026). The explicit de-risking trigger - converting fixed-cost coverage from cyclical commissions to recurring trail fees.

  "Strengthening the stable revenue base through growth in investment-trust balances and recommended-stock assets" - recurring management framing across FY2025-FY2026 briefings.

• ROE of 8% or higher as a medium-term target (medium-term plan, 26 Apr 2024), explicitly tied to the TSE "cost-of-capital and share-price-conscious management" initiative - implying continued capital return as a lever.

• Continuation of special dividends through FY2028 (reiterated at FY2026 results, 28 Apr 2026, and the interim-dividend notice, 16 Sep 2025) - a committed, multi-year capital-return trigger layered on the ≥50% ordinary-payout policy.

• Digital "investment navigation" / MARUSAN-NET build-out (FY2025-FY2026 briefings) - online fund-selection and advisory tools to lift salesforce productivity and reach clients between branch visits, supporting rather than replacing face-to-face.

• Branch-network optimisation (e.g. branch relocation announced 16 Sep 2025) - repositioning the 29-location footprint toward more productive sites.

• Group simplification via the Marusan Finance merger (effective 1 Apr 2025) - consolidating the financing function into the parent.

Note: Marusan deliberately does not publish a numerical earnings forecast for the next fiscal year (it cited market uncertainty when declining FY2027 guidance at the FY2026 results), so all forward triggers are strategic/AUM targets rather than P&L guidance.

Section 8: Key risks

• Market-cycle dependence (high probability, high amplitude). Brokerage and solicitation commissions - the majority of revenue alongside trail fees - swing directly with Japanese retail trading appetite. The FY2025 brokerage line fell ~21% in a soft tape, then rose ~36% in the buoyant FY2026 tape. A sustained Japanese equity downturn would compress two of three revenue pillars simultaneously and shrink AUM (and therefore trail fees) at the same time. This is the dominant risk and management acknowledges the revenue structure's "high sensitivity to stock-market conditions, interest rates and exchange movements."

• Structural fee compression from online brokers (high probability, slow drag). SBI, Rakuten and peers have driven equity-brokerage and front-load fund fees toward zero. Marusan's transactional pillars are priced well above the online standard; it defends them with advice, but the secular direction is down. The mechanism is gradual: each year, a little more of the new-investor flow routes to free platforms, and fiduciary-conduct pressure makes front-load selling harder to justify.

• Demographic erosion of the client base (moderate probability, structural, long-fuse). Marusan's franchise is built on older, advice-seeking, asset-rich households. That cohort is, by definition, ageing. Assets transfer at death to a generation more comfortable self-directing online. The trail-fee book - the firm's prize - is only as durable as the clients who hold it. This is the risk that no amount of branch optimisation fixes.

• The recurring-revenue project could fall short (moderate probability). The entire de-risking thesis rests on growing trail fees to cover 55% of SG&A by FY2028; at the interim stage it stood at 49%. If net fund inflows slow or a market drawdown shrinks AUM, the coverage ratio stalls and Marusan remains as cyclical as ever - exactly when it most needs the ballast.

• Payout sustainability (moderate probability, self-inflicted). Dividends have been paid at payout ratios above 100% of earnings in several recent years (FY2024 ~135%, FY2026 ~105%), funded by a fortress balance sheet (~66% equity ratio, net cash). This is sustainable for now because the balance sheet is over-capitalised, but it is a return of capital as much as a return on it; a multi-year earnings trough plus a >100% payout would draw down the buffer that is itself part of the client-trust proposition.

• Fiduciary-conduct / regulatory tightening (moderate probability). The FSA's "customer-first" push directly targets commission-driven advised sales. Tighter rules on front-load fees or churn would hit the solicitation line and constrain the salesforce's revenue model.

• No forward guidance (transparency, not solvency, risk). Marusan declines to publish next-year earnings forecasts. This is normal for Japanese brokers but leaves shareholders with less visibility and makes the stock a blunter proxy for the Nikkei than a guided industrial.

Section 9: Walk the talk

Five reporting periods used: FY2025 full-year (28 Apr 2025), Q1 FY2026 (30 Jul 2025), H1 FY2026 (~30 Oct 2025), Q3 FY2026 (~30 Jan 2026), FY2026 full-year (28 Apr 2026). No transcripts exist; this assessment cross-references the Tanshin filings, the strategy briefings, and the April 2024 medium-term plan against outcomes.

The anchor commitment is the medium-term plan of 26 April 2024: grow equity-fund net inflows by ¥300bn and recommended-stock assets by ¥100bn by FY2028, lift trust-fee coverage of SG&A toward 55%, target ROE ≥8%, and return capital via a ≥50% payout plus special dividends through FY2028. Everything since can be judged against that scorecard.

On growing the recurring book, management has tracked ahead of its own pace. By the first interim checkpoint, equity investment-trust net inflows were running above plan and the recommended-Japanese-stock target was reported at well over its annualised goal (~168% of the 12-month pace). The trust-fee-to-SG&A coverage ratio reached 49.0% at the H1 FY2025 stage against a 55% FY2028 target - on trajectory, not yet delivered. This is the most credible part of the story: the AUM-growth promises are being measurably met, in a favourable market.

On the cyclical lines, management was honest about the swing rather than spinning it. The FY2025 results (28 Apr 2025) showed brokerage commissions down ~21% in a soft tape, and the firm reported it plainly. Then FY2026 (28 Apr 2026) delivered a sharp recovery - operating revenue +15.3%, ordinary income +44%, brokerage commissions +36.3%, trust fees +10.6% - which management attributed to the buoyant market rather than to structural transformation. The Q1 FY2026 filing (30 Jul 2025) is a useful credibility tell: operating income fell ~62% and ordinary income ~49% as transactional revenue softened, even as net income rose ~52% on a one-off securities gain. Management let the operating-line weakness show rather than leading with the flattering net-income figure - a point in favour of straight reporting.

On capital return, the talk and the walk match precisely. The dividend promises have been kept to the cent. Special dividends were reinstated from FY2024 as promised; the ≥50% payout policy has been honoured (in practice well above it); and at the FY2026 results management both raised the annual dividend to ¥70 and reaffirmed special dividends through FY2028.

"We intend to continue special dividends through the fiscal year ending March 2028, in addition to maintaining a payout ratio of 50% or more of consolidated net income" - reaffirmed at FY2026 results, 28 Apr 2026, consistent with the 26 Apr 2024 plan.

Where to be skeptical. Almost all of the delivery has occurred in a rising market. The genuine test of the "stability" thesis - whether the trail-fee book actually cushions earnings in a downturn - has not yet happened during the plan period. And management's refusal to guide next-year earnings means there is no near-term promise to hold them to; they commit to multi-year AUM targets and dividends, not to P&L outcomes.

Assessment: this is management that does roughly what it says, with the important caveat that it has only been tested in good weather. On the things fully within its control - capital return and AUM-gathering effort - it has been accurate and consistent, even slightly under-promising on dividends. It has reported cyclical weakness honestly rather than hiding it. The unproven claim is the structural one: that the firm is becoming meaningfully less cyclical. Credible, consistent, not yet stress-tested.

Section 10: Shareholder friendliness index

Dividends. Marusan has become markedly more generous. Dividend per share over the last three fiscal years: FY2024 ¥60 (¥40 ordinary + ¥20 special), FY2025 ¥60 (¥60 ordinary), and FY2026 ¥70 (¥40 ordinary + ¥30 special). That follows a sharp step-up from just ¥12 in FY2023, when the special-dividend programme was reinstated under the April 2024 medium-term plan and the TSE's cost-of-capital initiative. The policy is a floor of ≥50% of consolidated net income, with special dividends committed through FY2028. Payout ratios have repeatedly exceeded 100% of earnings (≈135% in FY2024, ≈105% in FY2026), which signals deliberate balance-sheet normalisation - Marusan is over-capitalised (equity ratio ~66%, net cash) and is returning excess capital, not stretching to fund the dividend from a thin cushion.

Buybacks and dilution. Per the MoatMap database block, there have been zero buybacks in the last ~90 days (since 15 Mar 2026). Searching the longer history: Marusan's last meaningful repurchases were older - roughly ¥0.5bn in FY2022 and ~¥0.56bn back in 2015 - and I found no on-market buyback programme executed in the FY2024-FY2026 window; capital return over the last three years has been delivered entirely through dividends. Treasury stock sits at roughly 1.8% of shares outstanding. The share count has been broadly stable (~67.3m shares as of March 2025) with no option-driven dilution of note - this is not a company creating shares. So over three years: dividends sharply up and well above policy, no recent buyback, share count flat.

Verdict: Returns Capital - a high, rising, multi-year-committed dividend funded out of an over-capitalised balance sheet, with a flat share count, even though buybacks have not been part of the recent toolkit.

Section 11: Insider activities

Source: Japan's TSE/EDINET insider-disclosure regime is API/portal-gated and not reliably reachable via web search, so per instruction the MoatMap cross-market database is the sole source for recent insider dealing here.

MoatMap records 0 insider transactions for 8613.T over the trailing 12 months (data current 2026-06-13). There are no director, officer, or substantial-shareholder open-market buys or sells in the window.

For context on the ownership backdrop (from public shareholder filings, not transaction filings): the register is anchored by Nippon Life Insurance (~7.9%), the founding-family-linked Nagao Natural Environment Foundation (~7.2%), and MUFG Trust Bank (~2.5%), with treasury stock ~1.8%. These are stable, long-standing holdings rather than recent transactions.

Net assessment: no insider transaction signal is available - neither buying nor selling in the last 12 months per the canonical source. The stable, concentrated register (a life insurer and a founding-family foundation among the top holders) suggests a quiet, closely-held shareholder base rather than active insider trading. Neutral - no signal, reflecting absence of recorded activity rather than evidence of insider conviction either way.

Section 12: Scenarios

Bull case. The "savings to investment" shift compounds. NISA inflows keep arriving, the Japanese equity market stays firm, and Marusan's advised model proves it can win the older, wealthier slice of the new-investor wave that online brokers can't fully serve. Net fund inflows hit and exceed the ¥300bn plan target; the trail-fee book grows large enough that recurring fees cover well over 55% of fixed costs, and for the first time a soft market quarter passes without earnings cratering - proof that the firm has genuinely de-risked. Rising domestic rates fatten financial income on the net-cash balance sheet. Management keeps over-paying dividends out of surplus capital while the franchise quietly grows, and the firm emerges as a consolidator of weaker independent peers, buying scale cheaply. The advised-retail niche turns out to be defensible after all, anchored by a sticky, trusting, asset-rich client base.

Base case. Marusan stays what it is: a well-capitalised, well-run, cyclical retail broker geared to the Japanese market. In good market years (like FY2026) brokerage and solicitation surge and profits jump; in soft years they sag, partly cushioned by the growing-but-not-yet-dominant trail-fee book. The medium-term AUM targets are broadly met because the market cooperates, and trust-fee coverage drifts toward the high-40s/low-50s percent - real progress, short of full transformation. The dividend stays high and is faithfully paid, special dividends run through FY2028 as promised, and the share count stays flat. The stock behaves as a leveraged, high-yielding proxy for Japanese retail risk appetite. Nothing breaks; nothing transforms.

Bear case. A sustained Japanese equity downturn hits all three of Marusan's pillars at once - brokerage and solicitation collapse with trading volumes, and falling markets shrink AUM so the supposedly stable trail-fee line drops too, exposing the de-risking thesis as untested theory. Meanwhile the secular grind continues underneath: online brokers and fiduciary-conduct pressure keep eroding the transactional fee pillars, and the ageing client base transfers assets to heirs who self-direct on free platforms. Earnings fall while the >100%-of-earnings dividend keeps drawing down the capital buffer that underpins client trust, forcing an eventual dividend cut that breaks the one promise management had reliably kept. Marusan shrinks into a slowly-fading regional advisory house - solvent, over-capitalised, but structurally in decline - until it becomes a target rather than a consolidator.

Sources

• Marusan Securities IR page and press releases
• Stock Analysis - 8613 company profile
• irbank - 8613 financial data and dividend history
• Kitaishihon - 8613 business detail
• Smart Stock Notes - Marusan Securities (8613) analysis
• Nikkei - 8613 medium-term plan disclosure (26 Apr 2024)
• Nikkei - 8613 interim dividend notice (16 Sep 2025)
• Nikkei - 8613 Q3 FY2026 flash (20 Jan 2026)
• 官報決算データベース - Q1 FY2026 Tanshin summary
• Strainer - Marusan company history and Japanese Wikipedia - 丸三証券
• SimplyWall.st - Marusan FY2025 earnings (EPS ¥57.38)
• MoatMap multiverse insider/buyback database (market JP), as of 2026-06-13

Section 13 (Further Reading) is omitted: SemiAnalysis, Stratechery, and MBI Deep Dives have no coverage of this company.

Should I save this as a .md file in the working directory? I'll write it on your confirmation, or to a path you specify.